List your company

Mathematical Reasoning on MATH 500, AIME 2024, AIME 2025, AMC 2023, and Olympiad Bench

76.36Average ScoreGemini-2.5-flash
16.913632.346847.7863.2132May 13, 2026May 16, 2026May 19, 2026May 22, 2026May 25, 2026May 28, 2026May 31, 2026

Evaluation results

MethodLinks
Gemini-2.5-flashStudent Model=-, Teacher Model=Gemini-2.5-flash, Method=Base2026.0576.36—————
OmniOPDStudent Model=Qwen3-4B, Teacher Model=Gemini-2.5-flash, Method=OmniOPD2026.0575.67—————
OmniOPDStudent Model=Qwen3-4B, Teacher Model=Claude-4.5-haiku, Method=OmniOPD2026.0574.92—————
SFTStudent Model=Qwen3-4B, Teacher Model=Gemini-2.5-flash, Method=SFT2026.0573.51—————
OmniOPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B-Inst., Method=OmniOPD2026.0572.32—————
Qwen3-32BStudent Model=-, Teacher Model=Qwen3-32B, Method=Base2026.0571.62—————
Qwen3-4B (GRPO)Student Model=Qwen3-4B, Teacher Model=-, Method=GRPO2026.0570.24—————
OmniOPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-32B, Method=OmniOPD2026.0569.08—————
SFTStudent Model=Qwen3-4B, Teacher Model=Claude-4.5-haiku, Method=SFT2026.0567.52—————
Claude-4.5-haikuStudent Model=-, Teacher Model=Claude-4.5-haiku, Method=Base2026.0566.03—————
OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-32B, Method=OPD2026.0564.16—————
SFTStudent Model=Qwen3-4B, Teacher Model=Qwen3-32B, Method=SFT2026.0563.8—————
QwQ-32B-Previewmode=zero-shot, decoding=greedy2026.0562.590.65033.377.561.2
Qwen3-1.7B (GRPO)Student Model=Qwen3-1.7B, Teacher Model=-, Method=GRPO2026.0562.41—————
Qwen3-30B-A3B-Inst.Student Model=-, Teacher Model=Qwen3-30B-A3B-Inst., Method=Base2026.0562.12—————
OPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B, Method=OPD2026.0561.7—————
OmniOPDStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B, Method=OmniOPD2026.0560.67—————
OPDStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B-Inst., Method=OPD2026.0556.22—————
Qwen3-4BStudent Model=Qwen3-4B, Teacher Model=-, Method=Base2026.0554.01—————
SFTStudent Model=Qwen3-1.7B, Teacher Model=Qwen3-32B, Method=SFT2026.0553.76—————
Qwen3-1.7BStudent Model=Qwen3-1.7B, Teacher Model=-, Method=Base2026.0551.87—————
STRIDE-Qwen-7Bmode=zero-shot, decoding=greedy2026.0551.184.626.723.37546.1
SFTStudent Model=Qwen3-4B, Teacher Model=Qwen3-30B-A3B-Inst., Method=SFT2026.0549.77—————
Eurus-2-7B-PRIMEmode=zero-shot, decoding=greedy2026.0544.880.426.713.36043.7
Qwen2.5-Math-7B-Instructmode=zero-shot, decoding=greedy2026.0542.983.616.71062.541.6
Qwen2.5-7B-Instructmode=zero-shot, decoding=greedy2026.053675.5106.752.535.5
STRIDE-Llama-8Bmode=zero-shot, decoding=greedy2026.053670.413.31050.336
OpenMath2-Llama3.1-8Bmode=zero-shot, decoding=greedy2026.0528.867.86.73.337.528.9
Llama-3.1-8B-Instructmode=zero-shot, decoding=greedy2026.0519.251.93.33.322.515.1
Claude3.5-Sonnetmode=zero-shot, decoding=greedy2026.05—78.316———
GPT-4omode=zero-shot, decoding=greedy2026.05—76.69.3—47.543.3
Llama-3.1-70B-Instructmode=zero-shot, decoding=greedy2026.05—6813.3—42.529.4
NuminaMath-72B-CoTmode=zero-shot, decoding=greedy2026.05—643.3—7032.6
o1-minimode=zero-shot, decoding=greedy2026.05—9056.7—9565.3
o1-previewmode=zero-shot, decoding=greedy2026.05—85.544.6—90—
OpenMath2-Llama3.1-70Bmode=zero-shot, decoding=greedy2026.05—71.813.3—4530.1
Qwen2.5-Math-72B-Instructmode=zero-shot, decoding=greedy2026.05—82.623.3—7049
rStar-Math-7Bmode=zero-shot, decoding=greedy2026.05—78.426.7—47.547.1